On Jan  1, 2011, Richard Stallman <[email protected]> wrote:

> it still seems to me that the directory which normally holds firmware
> files could effectively be that list.

If there was such a directory, yeah, it could be.  But there's no such
directory.  It's up to the hotplug script to decide where to look for
the firmware bits it's going to give back to the kernel.  It can be a
directory, a collection of directories, a tarball, a database, anything.
So the kernel can't look it up.

And then, if the kernel could look it up, it might as well get the data
straight from the filesystem.  The reason is doesn't is precisely to
enable userland to control not only where, but also how firmware is
stored.

> The function that requests loading firmware look in the firmware
> directory for the relevant file name.  If it finds a file, it requests
> loading the firmware.  Otherwise, if the firmware file is free, it
> requests loading the firmware.  But if the file is nonfree and not present,
> it gives an error message, "Device not supported".

> Wouldn't this do the job?

It would, if the kernel could tell how userland would resolve filename
requests.  It can't.  That's why my proposal involves userland to tell
the kernel what pieces of firmware are available upfront.

-- 
Alexandre Oliva, freedom fighter    http://FSFLA.org/~lxoliva/
You must be the change you wish to see in the world. -- Gandhi
Be Free! -- http://FSFLA.org/   FSF Latin America board member
Free Software Evangelist      Red Hat Brazil Compiler Engineer

Reply via email to